The empirical reason is that being a newsletter subscriber correlates with having spent about twice the recorded time on our website, compared to non-subscribers.
I feel like the causality could be “people spend a lot of time on the website” --> “people subscribe to the newsletter” rather than the other way?
And how does 80k record people’s time spent on the website?
In fact, I think it’d be really weird if there wasn’t at least some effect in that direction, because people have to go on our website at some point in order to sign up to the newsletter!
I do think there is a significant effect in the direction ‘people subscribe to the newsletter → people spend time on the website,’ though.
Why? Mostly cos of the big spikes in engagement we see on pages that we link to from the newsletter — it just does look like people click on and then sometimes read stuff on our site, if we put it in the newsletter.
As for how we record people’s time spent on the website: we use the web analytics tools Google Analytics and Mixpanel. There are various caveats and systematic measurement errors that apply to both of them that we try to account for in our reporting on web engagement statistics.
I feel like the causality could be “people spend a lot of time on the website” --> “people subscribe to the newsletter” rather than the other way?
And how does 80k record people’s time spent on the website?
I totally agree the causality could go that way.
In fact, I think it’d be really weird if there wasn’t at least some effect in that direction, because people have to go on our website at some point in order to sign up to the newsletter!
I do think there is a significant effect in the direction ‘people subscribe to the newsletter → people spend time on the website,’ though.
Why? Mostly cos of the big spikes in engagement we see on pages that we link to from the newsletter — it just does look like people click on and then sometimes read stuff on our site, if we put it in the newsletter.
As for how we record people’s time spent on the website: we use the web analytics tools Google Analytics and Mixpanel. There are various caveats and systematic measurement errors that apply to both of them that we try to account for in our reporting on web engagement statistics.